Methanogen lyase homologous with high-dimensional characteristics of PeiP lyase and application of methanogen lyase

Abstract

The invention discloses methanogen lyase homologous with high-dimensional characteristics of PeiP lyase and application of the methanogen lyase, and aims to solve the problem that PeiP lyase protein of methanogen cell wall peptide bonds can participate in the biological process of hydrolyzing archaea cell walls, effectively kill methanogen and reduce methane production, and protein homologous with the protein has potential methane reduction potential. According to the method, a series of bioinformatics software is utilized, a protein sequence coded by methanogens virus is identified from a large amount of rumen microorganism virus group sequencing data, the high-dimensional homology of a deep learning model TM-Vec between the protein sequence and PeiP is analyzed, and the methanogens lyase homologous with the PeiP on the high-dimensional feature level of the model is obtained. According to the invention, multiple segments of primers are designed to synthesize a target sequence, the expression of the lyase is successfully realized in a prokaryotic expression system, and the effectiveness of the novel lyase in the aspect of reducing the yield of methane is verified in an in-vitro gas production test by detecting a crude enzyme solution.

Description

Technical Field

[0001] The present invention relates to the fields of protein function prediction and genetic engineering, and in particular to a methanogenic bacteria lyase having high-dimensional characteristics homologous to PeiP lyase and an application thereof. Background Art

[0002] Methane is an important factor affecting global climate change. Ruminants produce a large amount of methane during the digestion process. This process is mainly attributed to the degradation of plant cell wall components (such as cellulose and hemicellulose) by rumen microorganisms, producing intermediate metabolites such as volatile fatty acids, carbon dioxide (CO2) and hydrogen, which are then used by methanogens to synthesize methane.

[0003] Methanogens belong to archaea, which are quite different from bacteria in phylogeny, and their cell wall structure is also different from that of bacteria. Among them, pseudopeptidoglycan is the main component of the cell wall of some methanogens (mainly found in Methanobacteriales and Methanopyrales), and its three-dimensional structure is similar to bacterial peptidoglycan, but it has the following significant characteristics:

[0004] 1. The polysaccharide backbone of pseudopeptidoglycan is composed of N-acetylaminotaluronic acid, which is unique to archaea.

[0005] 2. It is connected to N-acetylglucosamine or N-acetylgalactosamine through a β-1,3-glycosidic bond, which is different from the β-1,4-glycosidic bond of bacterial peptidoglycan.

[0006] 3. The peptide chain lacks D-amino acids, and ε- and γ-isopeptide bonds are used in peptide-peptide cross-linking, which further increases its stability.

[0007] It is these structural features that make methanogens containing pseudopeptidoglycan layers highly resistant to lysozyme and other bacterial cell wall hydrolases. Therefore, the development of enzymes that can specifically hydrolyze pseudopeptidoglycan is of great significance for methane inhibition strategies.

[0008] Previous studies have identified a methanogen lyase PeiP in the prophage ΨM2 of Methanothermobacter marburgensis, which can hydrolyze pseudopeptidoglycan (Reference: Pseudomurein endoisopeptidases PeiW and PeiP, two moderately related members of a novel family of proteases produced in Methanothermobacter strains). However, PeiP is prone to precipitation at physiological temperature, has a short half-life in vivo, and poor structural stability, which is not conducive to practical applications. Therefore, the present invention screens potential peptidase proteins from rumen microbiome data and evaluates their similarity with PeiP through deep learning homology search, and a novel methanogen lyase is mined therefrom. Subsequently, its function is characterized by heterologous expression, and its effect of inhibiting methane production is verified in vitro. The research results of the present invention are expected to reduce methane emissions in the livestock industry and improve feed conversion efficiency at the same time, which is of great significance for promoting the green development of the livestock industry. Summary of the Invention

[0009] The present invention aims to solve the problems of high methane emissions and lack of green and efficient methane inhibitors in the livestock industry (especially ruminant production). By means of the growing microbial genome sequencing data resources, combined with the cutting-edge technologies in the fields of deep learning and bioinformatics, through systematic data mining, a methanogen lyase Pei277 homologous to the high-dimensional features of the PeiP lyase is screened. After prokaryotic system expression, the recombinant enzyme exhibits good methane inhibition ability.

[0010] The object of the present invention is achieved by the following technical solutions: a methanogen lyase homologous to the high-dimensional features of the PeiP lyase, and the amino acid sequence of the lyase is shown in SEQ ID NO.1.

[0011] Furthermore, the process of screening the lyase is as follows:

[0012] (1) Collection of rumen metagenomic data and mining of archaeal virus proteins

[0013] Collect publicly available rumen metagenomic assemblies, metagenome-assembled genomes, and virome data. Use the virus identification software geNomad to mine viral genomes, use CheckV to remove multi-host contamination, incomplete viral genomes, and trim host sequence contamination, and perform dereplication at the species level with an average nucleotide similarity of 95% and an alignment coverage of 85% to obtain vOTUs. Use iPhoP to predict the potential hosts of vOTUs and retain viral genomes that infect archaea of the order Methanobacteriales. Further, use prodigal-gv to predict the encoded proteins; use the Merops peptidase database to annotate whether the obtained proteins belong to peptidases and remove the non-peptidase parts.

[0014] (2) Screening by high-dimensional feature homology alignment

[0015] Use the TM-Vec deep learning homology model to determine whether the proteins obtained in step (1) are homologous to PeiR in terms of high-dimensional features of the model, and obtain a methanogen lyase homologous to PeiR at the high-dimensional feature level of the Tm-Vec model according to the set TM-Score threshold.

[0016] Furthermore, the specific process of determining whether a protein belongs to a peptidase is as follows: Based on the target substrate being the peptide chain between sugar backbones, the obtained viral proteins are annotated by Diamond blastp alignment against the Merops peptidase database, and the protein sequences that are peptidases are retained for subsequent analysis.

[0017] Furthermore, use the ProtParam module in the Biopython SeqUtils package to predict the stability of the obtained proteins for the methanogen lyase.

[0018] Furthermore, use the deep learning model TM-Vec that links sequence and structure levels and allows searching for structure-structure similarity in large sequence libraries to perform homology alignment. By calculating the TM-score, determine whether the target protein has a homologous relationship with the template protein PeiP in the high-dimensional feature space. Select TM-score ≥ 0.6 as the judgment standard threshold for reliable homology, and proteins that meet this threshold are considered homologous to PeiP.

[0019] Furthermore, based on the recombinant vector and recombinant bacteria containing the methanogen lyase gene, perform fermentation-induced expression, and obtain the target protein through subsequent purification, and evaluate the effect of the target protein on in vitro microbial methane production.

[0020] On the other hand, the present invention also provides an application of a methanogen lyase that is highly homologous to the PeiP lyase in inhibiting methane production.

[0021] Advantages of the present invention: The present invention widely collects a large amount of rumen microbial virome sequencing data, uses a series of bioinformatics software to identify the protein sequences encoded by methanogen viruses from it, and on this basis analyzes the high-dimensional homology of these proteins with TM-Vec, a deep learning model between PeiP. Finally, the present invention successfully obtains a novel methanogen lyase that is homologous to PeiP at the high-dimensional feature level of the model. To achieve the expression of this lyase, the present invention designs multiple primers to synthesize the target sequence and successfully expresses it in a prokaryotic expression system. Through the detection of the crude enzyme solution, we verified the effectiveness of this novel lyase in reducing methane production in an in vitro gas production experiment. This novel methanogen lyase provides a new strategy and tool for methane emission reduction in the livestock industry and has important research and application prospects. Brief Description of the Drawings

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0023] Figure 1 is the PCR product of cloning methanogen lyase;

[0024] Figure 2 is the SDS-PAGE analysis chart of the crude enzyme solution of the recombinant protein;

[0025] Figure 3 is the effect of the recombinant protein on methane production. Detailed Embodiments

[0026] The following further details the present invention by combining specific implementation schemes. The following implementation schemes will help those skilled in the art further understand the present invention, but do not limit the present invention in any form. It should be noted that for those of ordinary skill in the art, any form of change and deformation made to the present invention without departing from the concept of the present invention belongs to the protection scope of the present invention.

[0027] The present invention provides a methanogen lyase homologous to the high-dimensional features of the PeiP lyase and its mining and characterization methods, and analyzes its application effect in inhibiting methane production, including the processes of microbiome data mining, sequence homology search, gene synthesis, protein expression, and methane inhibition effect evaluation. The specific steps are as follows:

[0028] 1. Microbiome data mining

[0029] Collect relevant literature on rumen microbiome and virome, download the corresponding metagenomic assembly data, metagenome-assembled genome data, and viral genome data, and add the assembly results of self-tested rumen metagenomic data. Input them into the software geNomad, which can combine gene content and deep neural network information for virus judgment, to mine and confirm the viral genomes therein. Use the CheckV software to judge the integrity and contamination of the viruses to remove low-quality genomes. At the same time, trim the potential host sequence contamination at both ends of the viral sequences to improve the accuracy of protein-coding function identification. And use 95% average nucleotide similarity and 85% alignment coverage as thresholds to perform dereplication at the species level to obtain the rumen viral genome database of ruminants. Further, use the virus-host relationship prediction tool iPhoP, and add 791 non-redundant rumen archaeal genomes to its original reference host genome library as candidate hosts to increase specificity. Predict virus hosts through means such as provirus alignment, CRISPR spacer alignment, and deep learning feature capture of viral genomes, and retain the viral genomes that infect methanogenic archaea of the order Methanobacteriales. Further, use the meta mode of the prodigal-gv software to predict the encoded viral protein set. Based on the target substrate being the peptide chain between the sugar backbones in the archaeal cell wall, align the obtained viral proteins with the Merops peptidase database by Diamondblastp, and retain the protein sequences that are peptidases for subsequent analysis.

[0030] 2. High-dimensional homology search of the deep learning homology model TM-Vec

[0031] The high-dimensional homology search of the deep learning homology model TM-Vec refers to searching for structural similarity in a large sequence library by combining protein sequence and structure information using the TM-Vec (1.0.2) model. The TM-Vec software uses a siamese neural network to predict the TM-Score of protein pairs as an indicator of structural similarity, generates protein vectors that can be effectively indexed and queried, without the need for intermediate structure calculations, generates structure-aware vector embeddings for protein sequences, and quickly searches for protein structures by finding the nearest neighbors in the embedding space, thus effectively narrowing the gap between protein sequences and structure information, and providing substantial improvements in terms of speed and sensitivity, with significantly higher efficiency than homology alignment based on predicted structures. Proteins that are homologous to PeiP at the high-dimensional level (TM-Score ≥ 0.6) are classified as PeiP homologous proteins and regarded as novel methanogen lyases. Predict the stability of these homologous proteins through the ProtParam module in the Biopython SeqUtils package. If the instability_index ≤ 40, then it is considered that the protein can exist stably.

[0032] One novel methanogen lyase obtained in the present invention is homologous to PeiP at the high-dimensional feature level of the deep learning model and can stably exist, as shown in the following table:

[0033]

[0034] 3. Heterologous expression of lyase

[0035] (1) Synthesis of lyase gene and construction of recombinant plasmid

[0036] The gene sequence of the target protein was designed in segments, and multiple segments were assembled into a complete target sequence by Overlap PCR. The pET-30a(+) vector was double-digested and linearized with XhoⅠ and NdeⅠ restriction endonucleases, and the target DNA fragment was ligated to the linearized vector using seamless cloning technology. It was then transformed into Escherichia coli DH5α competent cells, and positive clones were obtained by screening with kanamycin. After amplification culture, they were preserved.

[0037] (2) Transformation of recombinant plasmid and construction of expression strain

[0038] A high-quality recombinant plasmid Pei_pET-30a(+) was extracted using a plasmid miniprep kit and introduced into Escherichia coli BL21(DE3) competent cells by heat shock method. After recovery growth, positive clones were obtained by screening with kanamycin. As Figure 1 shown, after verifying the correctness of the gene sequence by colony PCR identification and Sanger sequencing, the engineering strain was preserved.

[0039] (3) Induced expression of recombinant protein

[0040] The verified engineering strain was inoculated into LB medium containing kanamycin for amplification culture. When the OD 600 of the bacterial solution reached 0.6, isopropyl β-D-1-thiogalactopyranoside (IPTG) with a final concentration of 0.25 mM was added, and induced expression was carried out at 16 °C for 18 hours. After centrifuging to collect the bacteria, lysis buffer was added to resuspend and remove cell debris to obtain a crude enzyme solution. Protein expression was verified by SDS-PAGE electrophoresis combined with Coomassie Brilliant Blue staining, and the results were as Figure 2 shown.

[0041] 4. Evaluation of the effect of recombinant protein on in vitro microbial methane production

[0042] The in vitro gas production technique was adopted to evaluate the effect of recombinant protein on microbial methane production. The rumen fluid of three dairy cows was extracted by a vacuum pump, mixed, filtered through four layers of gauze, and then injected into a gas production bottle at an addition ratio of 5 mL:45 mL with artificial saliva as the culture substrate; the fermentation substrate was TMR diet (collected from the pasture), and 500 mg of dry matter feed was added to each gas production bottle. 1 mL of crude enzyme solution was added to the treatment group, and the treatment was repeated 10 times; the control group (CON) was not added with recombinant protein, and 1 mL of pure water was used as the negative control. When culturing in an incubator at 39 °C for 12 h, 24 h, and 48 h, the pressure in the gas production bottle was read using a pressure sensor, and the gas was collected.

[0043] Gas production: calculation formula

[0044] GP t is the gas production (mL) of the sample at time t; P t is the pressure (mPa) read at time t; V0 is the volume of the bottle; 101.3 is the standard atmospheric pressure (mPa); W is the dry matter weight of the sample. The total cumulative gas production during the gas production process is the sum of the gas production at each time period.

[0045] Methane production: The methane content of the collected gas was measured using a gas chromatograph. Methane production = gas production × methane content. The results showed ( Figure 3 ): The protein Pei277 could significantly reduce the methane production of rumen microorganisms.

[0046] The above embodiments are used to explain and illustrate the present invention, rather than limiting the present invention. Any modifications and changes made within the spirit and scope of the protection of the claims of the present invention fall within the protection scope of the present invention.

Claims

1. A methanogen lyase homologous to the high-dimensional features of PeiP lyase, characterized in that The amino acid sequence of the lyase is shown in SEQ ID NO.

1.

2. A methanogen lyase having high-dimensional homology to PeiP lyase according to claim 1, characterized in that: The lytic enzyme screening process is as follows: (1) Rumen metagenomic data collection and archaeal viral protein mining Publicly published rumen metagenome assemblies, metagenome-assembled genomes, and virome data were collected. Viral genomes were mined using the virus identification software geNomad. Viral genomes with multiple host contamination and incomplete genomes were removed using CheckV, and host sequence contamination was removed. The average nucleotide similarity of 95% and the alignment coverage of 85% were used as thresholds for species-level redundancy removal to obtain vOTUs. iPhoP was used to predict potential hosts of vOTUs and viral genomes that infected Methanobacteriales archaea were retained. Prodigal-gv was further used to predict the encoded proteins. The Merops peptidase database was used to annotate whether the obtained proteins belonged to peptidases and to remove the non-peptidase parts. (2) High-dimensional feature homology comparison and screening The TM-Vec deep learning homology model is used to determine whether the protein obtained in step (1) has homology with PeiR in terms of the high-dimensional features of the model, and a methanogenic bacterial lyase that is homologous to PeiR in the high-dimensional features of the Tm-Vec model is obtained according to the set TM-Score threshold.

3. A methanogen lyase having high-dimensional homology to PeiP lyase according to claim 2, characterized in that: The specific process of determining whether a protein is a peptidase is as follows: based on the target substrate being a peptide chain between sugar backbones, the obtained viral protein is compared with the Merops peptidase database annotation by Diamond blastp, and the protein sequence that is a peptidase is retained for subsequent analysis.

4. A methanogen lyase having high-dimensional homology to PeiP lyase according to claim 2, characterized in that: The ProtParam module in the BiopythonSeq Utils package was used to predict the stability of the resulting protein for methanogen lyase.

5. A methanogen lyase having high-dimensional homology to PeiP lyase according to claim 2, characterized in that: Homology comparison was performed using TM-Vec, a deep learning model that links sequence and structure levels and allows searching for structure-structure similarities in large sequence libraries. TM-score was calculated to determine whether the target protein had a homology relationship with the template protein PeiP in the high-dimensional feature space. TM-score ≥ 0.6 was selected as the threshold for judging credible homology, and proteins that met this threshold were considered to be homologous to PeiP.

6. A methanogen lyase having high-dimensional homology to PeiP lyase according to claim 1, characterized in that: Based on the recombinant vector and recombinant bacteria containing the methanogen lyase gene, fermentation-induced expression was carried out, and the target protein was obtained through subsequent purification, and the effect of the target protein on in vitro microbial methane production was evaluated.

7. Application of a methanogenic bacterial lyase with high-dimensional features homologous to PeiP lyase in inhibiting methane production.

Images